An automatic driving network security situation awareness and emergency processing method and system

Abstract

The application discloses an automatic driving network security situation awareness and emergency processing method and system, relates to the technical field of intelligent driving, and specifically comprises the following steps: collecting multivariate data of a first situation element for reflecting a driving state, a second situation element for reflecting a driving state, and a third situation element for reflecting an environmental state in real time; constructing a risk field model based on the collected situation element data; superimposing each risk field to construct a comprehensive risk field model; calculating a safety risk assessment value at the current moment according to the collected multivariate situation element data; setting a risk threshold value; and when the safety risk assessment value is greater than the risk threshold value, performing a driving decision process, wherein the driving decision process makes the safety risk assessment value drop to a level less than the risk threshold value within a time interval according to the current state through one or more decision combination strategies, can dynamically adjust the driving decision strategy according to real-time state data, and improves the intelligence and real-time performance of the driving decision.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, and in particular to a method and system for cybersecurity situational awareness and emergency response in autonomous driving. Background Technology

[0002] With the rapid development of autonomous driving technology, autonomous vehicles are gradually becoming an important part of future transportation. Autonomous vehicles achieve autonomous driving functions through complex sensor networks, communication systems, and control systems, greatly improving traffic efficiency and safety. However, the digitalization and networking of autonomous vehicles also expose them to numerous safety risks.

[0003] Autonomous vehicles rely on numerous sensors (such as cameras, LiDAR, and millimeter-wave radar) to perceive their surroundings. The data generated during operation is diverse, including sensor data, communication data, and vehicle operational data. Existing systems' situational awareness and emergency response mechanisms often suffer from delays, failing to provide real-time responses to varying environmental conditions and diverse influencing factors. They lack intelligent decision-making capabilities and cannot dynamically adjust driving strategies based on real-time status data. Therefore, this paper proposes a method and system for cybersecurity situational awareness and emergency response in autonomous driving systems. Summary of the Invention

[0004] The main objective of this invention is to provide a method and system for network security situation awareness and emergency response for autonomous driving, which can effectively solve the problems in the background art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A method for cybersecurity situational awareness and emergency response in autonomous driving includes:

[0007] Real-time acquisition of multi-dimensional situational awareness data, wherein the situational awareness data includes:

[0008] At least one first situational element used to reflect the driving status;

[0009] At least one second situational element used to reflect the driving status;

[0010] And at least one third situational element used to reflect the environmental state;

[0011] A risk field model is constructed based on the collected situational data, wherein the risk field includes:

[0012] First risk field model based on first situation element data ;

[0013] Second risk field model based on second situation element data ;

[0014] And, the third risk field model based on the third situation element data. ;

[0015] By overlaying the various risk fields, a comprehensive risk field model is constructed. Calculate the current moment based on the collected multi-dimensional situational data. Security risk assessment value ;

[0016] Set risk threshold When the safety risk assessment value Greater than the risk threshold During this process, a driving decision-making process is conducted, which, based on the current state, uses one or more decision-making combinations to adjust the safety risk assessment value. In time interval The internal temperature dropped below the risk threshold. The level, wherein the decision includes:

[0017] At least one first driving decision used to change the first situational element;

[0018] At least one second driving decision used to alter the second situational element;

[0019] And at least one third driving decision used to change the third situational element.

[0020] An autonomous driving cybersecurity situational awareness and emergency response system includes:

[0021] The data acquisition module is used to collect multi-dimensional situational awareness data in real time, including:

[0022] The first data acquisition submodule is used to collect the first situational elements reflecting the driving status;

[0023] The second data acquisition submodule is used to collect the second situational element reflecting the driving status;

[0024] The third data acquisition submodule is used to collect the third situational element reflecting the environmental state;

[0025] The risk model building module is used to construct risk field models based on collected situational data, including:

[0026] Used to construct a first risk field model based on first situation element data. The first risk field model construction submodule;

[0027] Used to construct a second risk field model based on second situation element data. The second risk field model construction submodule;

[0028] Used to construct a third risk field model based on third situation element data. The third risk field model construction submodule;

[0029] The integrated risk field model construction module is used to overlay various risk fields to construct an integrated risk field model. ;

[0030] The security risk assessment module is used to calculate the current moment based on the collected multi-faceted situational data. Security risk assessment value ;

[0031] The driving decision-making module is used to determine the driving risk assessment value. Greater than the set risk threshold During this process, a driving decision-making process is conducted, wherein the driving decision-making process, based on the current state, employs one or more decision-making combinations to adjust the safety risk assessment value. In time interval The internal temperature dropped below the risk threshold. The levels include:

[0032] The first driving decision generation submodule is used to formulate changes to the first situational element;

[0033] The second driving decision generation submodule is used to formulate changes to the second situational factors;

[0034] The third driving decision generation submodule is used to formulate changes to the third situational element;

[0035] The driving decision execution module is used to execute the driving decision strategy generated by the driving decision formulation module.

[0036] The system also includes a memory, a processor, and a computer program stored in the memory and executable on the processor.

[0037] Furthermore, the first risk field model is a function of the driving state, and its expression is: = ;in, A non-zero constant correction coefficient; This represents the functional relationship between the first situational element and the driving state risk, reflecting the impact of driving state on the safety risk assessment value. The impact, This is the first situational data collected;

[0038] The second risk field model is a function of the driving state, and its expression is: = ;in, A non-zero constant correction coefficient; This represents the functional relationship between the second situational element and the driving state risk, reflecting the impact of the driving state on the safety risk assessment value. The impact, This is the second situational awareness data collected;

[0039] The third risk field model is a function of the environmental state, and its expression is: = ;in, A non-zero constant correction coefficient; This represents the functional relationship between the third situational element and environmental state risk, reflecting the impact of environmental state on the safety risk assessment value. The impact, This is the second situational awareness data collected;

[0040] The comprehensive risk field model is a superposition of various risk field models, and the expression for the superposition state is: = + + ;

[0041] The security risk assessment value is at the current time. The functional relationship between multi-faceted situational data and state risk is calculated using the following formula: = + + .

[0042] Furthermore, the first Functional relationship between situational factors and state risk The acquisition process includes the following steps:

[0043] Will with the The risk levels corresponding to the situation elements are classified into low risk, medium risk, and high risk according to the principle of low to high, and the risk values ​​of low risk, medium risk, and high risk are defined respectively. , and , and 0 < < < <1; =1,2,3;

[0044] Mark several different levels respectively Situational element data;

[0045] Construct the labeled first A neural network model is used as input, with the risk value corresponding to the data as output. The neural network model is trained until its classification accuracy reaches the set expected value.

[0046] Use the trained neural network model to obtain all the first The risk value of the state risk corresponding to the situation element is output, and the first risk value is output. The functional relationship between situational elements and their corresponding risk values ​​is used as the first... Functional relationship between situational factors and state risk .

[0047] Furthermore, the neural network model is a multilayer perceptron neural network model with an input layer, at least one hidden layer, and an output layer. The number of neurons in the input layer of the neural network model is equal to the number of input situational elements; the number of neurons in the output layer is 1; and the number of neurons in each hidden layer is not less than the average number of neurons in the input layer and the output layer.

[0048] Furthermore, the driving decision-making process includes the following steps:

[0049] Determine the current driving status, driving status, and environmental status based on situational data.

[0050] Based on the current status, select those that can achieve the security risk assessment value. Decreasing decision types, constructing an executable decision space ;

[0051] Construct a decision model, wherein the decision model is based on time intervals. Internally, through a series of decisions ={ , ,..., },and ∈ The optimization objective is to minimize the decision value function, where... For the initial decision; For the first One decision; the expression for the optimization objective is: In the formula, for The safety risk assessment value at any given time; the expression for the decision value function is: = In the formula, for The decision value function at any given time; for The decision value function at any given time; Discount factor;

[0052] The time interval is determined based on the optimization objective and the decision value function. A strategy that combines one or more decision-making processes.

[0053] Furthermore, the initial decision The determination process includes the following steps:

[0054] Acquire situational data reflecting the current state, and calculate the first risk value reflecting the risk of the driving state based on the situational data. The second risk value reflects the risk of driving conditions. and the third risk value reflecting the environmental status risk. ,in, = ; = ; = ;

[0055] Calculate the current time respectively Next, the Risk Value For safety risk assessment values Contribution ;in, = ; =1,2,3;

[0056] Take the maximum contribution The corresponding number Driving decisions as the initial decision space , where the initial decision space It contains several values ​​that enable security risk assessment. Decreasing decision types;

[0057] Take the initial decision space In the middle, make The decision to take the minimum value as the initial decision .

[0058] The present invention has the following beneficial effects:

[0059] Compared with existing technologies, this method involves real-time acquisition of at least one first situational element reflecting the driving state; at least one second situational element reflecting the driving state; and at least one third situational element reflecting the environmental state; and constructing a first risk field model based on the acquired situational element data. The second risk field model based on the second situation element data And the third risk field model based on the third situation element data By superimposing the various risk fields, a comprehensive risk field model is constructed. Calculate the current moment based on the collected multi-dimensional situational data. Security risk assessment value Set risk thresholds When the safety risk assessment value Greater than the risk threshold During this process, a driving decision-making process is conducted, which, based on the current state, uses one or more decision-making combinations to adjust the safety risk assessment value. In time interval The internal temperature dropped below the risk threshold. The system is capable of responding to environmental conditions and various influencing factors in real time, dynamically adjusting driving decision-making strategies based on real-time status data, and improving the intelligence and real-time nature of driving decisions. Attached Figure Description

[0060] Figure 1 This is a flowchart illustrating the autonomous driving network security situation awareness and emergency response method of the present invention.

[0061] Figure 2 This is a schematic diagram of the autonomous driving network security situation awareness and emergency response system of the present invention. Detailed Implementation

[0062] The present invention will be further described below with reference to specific embodiments. The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the present invention. In order to better illustrate the specific embodiments of the present invention, some parts in the drawings may be omitted, enlarged or reduced, and do not represent the actual product size.

[0063] The specific implementation process of the technical solution of this invention includes the following steps:

[0064] Step 1: Collect multi-dimensional situational data in real time.

[0065] Among them, the situational elements include:

[0066] At least one first situational element used to reflect the driving status;

[0067] At least one second situational element used to reflect the driving status;

[0068] And at least one third situational element used to reflect the environmental state;

[0069] Specifically,

[0070] The first situational element reflecting the driving state can be:

[0071] Physiological signals: Through devices such as physiological information monitoring wristbands, the driver's respiratory rate, heart rate and other physiological signals are collected in real time, and combined with preset thresholds to determine whether the driver is in an adverse state such as fatigue or sudden illness.

[0072] Behavioral and Facial Signals: High-definition and infrared cameras are used to collect multimodal parameters of the driver, including facial expressions, head posture, and torso posture. Deep learning algorithms, such as multi-task convolutional neural networks, are used to perceive the driver's fatigue and emotional state. The eye-tracking glasses' built-in camera captures real-time images of the scene and gaze points within the driver's field of vision. Object detection algorithms (such as YOLO) are used to identify objects within the scene images, and the recognition results are compared with the gaze points to determine if the driver is exhibiting distracted behavior.

[0073] Operational behavior signals, such as the frequency of steering wheel operation and the frequency of use of the brake and accelerator pedals, can be collected by setting up sensors.

[0074] The second situational element used to reflect the driving status can be:

[0075] Vehicle operating parameters: The vehicle's operating status parameters, such as speed, acceleration, and steering angle, are collected through sensors inside the vehicle, such as GPS locators, speed sensors, acceleration sensors, and steering wheel angle sensors.

[0076] Vehicle Surrounding Environment: Utilizing external high-definition cameras and onboard LiDAR, the system detects road environment information and nearby vehicle information. Deep neural networks (such as YOLOv7) process the collected external image information and radar point cloud information to identify and extract driving risk factors.

[0077] The third situational element reflecting the environmental state can be:

[0078] Scene recognition and risk assessment: Road environment information is input into a specific model, such as the SegmentAnything Model, for pixel-level segmentation and Grounding Dino to distinguish between pedestrian walkways and lanes. Based on the obtained position coordinates of pedestrians, pedestrian walkways, and lanes, the position of pedestrians in the scene is determined, thereby assessing potential risks.

[0079] Dynamic modeling of complex scenarios: In autonomous driving scenarios, the driving risk field model (DRF) is combined to dynamically adjust the driving risk field according to real-time road conditions and vehicle status in order to better predict and avoid potential risks.

[0080] Step 2: Construct a risk field model based on the collected situational data, including:

[0081] First risk field model based on first situation element data ;

[0082] Second risk field model based on second situation element data ;

[0083] And, the third risk field model based on the third situation element data. .

[0084] Specifically,

[0085] The first risk field model is a function of the driving state, and its expression is: = ;in, A non-zero constant correction coefficient; This represents the functional relationship between the first situational element and the driving state risk, reflecting the impact of driving state on the safety risk assessment value. The impact, This is the first situational data collected;

[0086] The second risk field model is a function of the driving state, and its expression is: = ;in, A non-zero constant correction coefficient; This represents the functional relationship between the second situational element and the driving state risk, reflecting the impact of the driving state on the safety risk assessment value. The impact, This is the second situational awareness data collected;

[0087] The third risk field model is a function of the environmental state, and its expression is: = ;in, A non-zero constant correction coefficient; This represents the functional relationship between the third situational element and environmental state risk, reflecting the impact of environmental state on the safety risk assessment value. The impact, This is the second situational awareness data collected;

[0088] No. Functional relationship between situational factors and state risk The acquisition process includes the following steps:

[0089] Will with the The risk levels corresponding to the situation elements are classified into low risk, medium risk, and high risk according to the principle of low to high, and the risk values ​​of low risk, medium risk, and high risk are defined respectively. , and , and 0 < < < <1; =1,2,3;

[0090] Mark several different levels respectively Situational element data;

[0091] Construct the labeled first A neural network model takes situational element data as input and outputs the corresponding risk value. The neural network model is trained until its classification accuracy reaches a set expected value. The neural network model is a multilayer perceptron neural network model with one input layer, at least one hidden layer, and one output layer. The number of neurons in the input layer of the neural network model is equal to the number of types of situational elements input. The number of neurons in the output layer is 1. The number of neurons in each hidden layer is not less than the average number of neurons in the input layer and the output layer.

[0092] Use the trained neural network model to obtain all the first The risk value of the state risk corresponding to the situation element is output, and the first risk value is output. The functional relationship between situational elements and their corresponding risk values ​​is used as the first... Functional relationship between situational factors and state risk .

[0093] Step 3: Overlay the various risk fields to construct a comprehensive risk field model. .

[0094] The comprehensive risk field model is a superposition of various risk field models, where the expression for the superposition state is: = + + .

[0095] Step 4: Calculate the current time based on the collected multi-dimensional situational data. Security risk assessment value ;

[0096] Among them, the safety risk assessment value is at the current moment. The functional relationship between multi-faceted situational data and state risk is calculated using the following formula: = + + .

[0097] Step 5: Set risk thresholds When the safety risk assessment value Greater than the risk threshold During the driving decision-making process, based on the current state, one or more decision-making strategies are used to adjust the safety risk assessment value. In time interval The internal temperature dropped below the risk threshold. The level, wherein the decision includes: at least one first driving decision for changing a first situational element; at least one second driving decision for changing a second situational element; and at least one third driving decision for changing a third situational element.

[0098] Specifically, the driving decision-making process includes the following steps:

[0099] Determine the current driving status, driving status, and environmental status based on situational data.

[0100] Based on the current status, select those that can achieve the security risk assessment value. Decreasing decision types, constructing an executable decision space ;

[0101] Construct a decision model, wherein the decision model is based on time intervals. Internally, through a series of decisions ={ , ,..., },and ∈ The optimization objective is to minimize the decision value function, where... For the initial decision; For the first One decision; the expression for the optimization objective is: In the formula, for The safety risk assessment value at any given time; the expression for the decision value function is: = In the formula, for The decision value function at any given time; for The decision value function at any given time; Discount factor;

[0102] The time interval is determined based on the optimization objective and the decision value function. A strategy that combines one or more decision-making processes.

[0103] Among them, the initial decision The determination process includes the following steps:

[0104] Acquire situational data reflecting the current state, and calculate the first risk value reflecting the risk of the driving state based on the situational data. The second risk value reflects the risk of driving conditions. and the third risk value reflecting the environmental status risk. ,in, = ; = ; = ;

[0105] Calculate the current time respectively Next, the Risk Value For safety risk assessment values Contribution ;in, = ; =1,2,3;

[0106] Take the maximum contribution The corresponding number Driving decisions as the initial decision space , where the initial decision space It contains several values ​​that enable security risk assessment. Decreasing decision types;

[0107] Take the initial decision space In the middle, make The decision to take the minimum value as the initial decision .

[0108] The autonomous driving cybersecurity situational awareness and emergency response system of the present invention specifically includes:

[0109] The data acquisition module is used to collect multi-dimensional situational awareness data in real time, including:

[0110] The first data acquisition submodule is used to collect the first situational elements reflecting the driving status;

[0111] The second data acquisition submodule is used to collect the second situational element reflecting the driving status;

[0112] The third data acquisition submodule is used to collect the third situational element reflecting the environmental state;

[0113] The risk model building module is used to construct risk field models based on collected situational data, including:

[0114] Used to construct a first risk field model based on first situation element data. The first risk field model construction submodule;

[0115] Used to construct a second risk field model based on second situation element data. The second risk field model construction submodule;

[0116] Used to construct a third risk field model based on third situation element data. The third risk field model construction submodule;

[0117] The integrated risk field model construction module is used to overlay various risk fields to construct an integrated risk field model. ;

[0118] The security risk assessment module is used to calculate the current moment based on the collected multi-faceted situational data. Security risk assessment value ;

[0119] The driving decision-making module is used to determine the driving risk assessment value. Greater than the set risk threshold During this process, a driving decision-making process is conducted, wherein the driving decision-making process, based on the current state, employs one or more decision-making combinations to adjust the safety risk assessment value. In time interval The internal temperature dropped below the risk threshold. The levels include:

[0120] The first driving decision generation submodule is used to formulate changes to the first situational element;

[0121] The second driving decision generation submodule is used to formulate changes to the second situational factors;

[0122] The third driving decision generation submodule is used to formulate changes to the third situational element;

[0123] The driving decision execution module is used to execute the driving decision strategy generated by the driving decision formulation module.

[0124] The autonomous driving network security situation awareness and emergency response system of the present invention further includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the autonomous driving network security situation awareness and emergency response method of the present invention.

[0125] In summary, by real-time collection of at least one first situational element reflecting the driving state; at least one second situational element reflecting the driving state; and at least one third situational element reflecting the environmental state, a first risk field model based on the collected situational element data is constructed. The second risk field model based on the second situation element data And the third risk field model based on the third situation element data By superimposing the various risk fields, a comprehensive risk field model is constructed. Calculate the current moment based on the collected multi-dimensional situational data. Security risk assessment value Set risk thresholds When the safety risk assessment value Greater than the risk threshold During this process, a driving decision-making process is conducted, which, based on the current state, uses one or more decision-making combinations to adjust the safety risk assessment value. In time interval The internal temperature dropped below the risk threshold. The system is capable of responding to environmental conditions and various influencing factors in real time, dynamically adjusting driving decision-making strategies based on real-time status data, and improving the intelligence and real-time nature of driving decisions.

[0126] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for cybersecurity situational awareness and emergency response in autonomous driving, characterized in that, include: Real-time acquisition of multi-dimensional situational awareness data, wherein the situational awareness data includes: At least one first situational element used to reflect the driving status; At least one second situational element used to reflect the driving status; And at least one third situational element used to reflect the environmental state; A risk field model is constructed based on the collected situational data, wherein the risk field includes: The first risk field model R based on the first situation element data behavior (se 1 ); The second risk field model R based on the second situation element data status (se 2 ); And, the third risk field model R based on the third situation element data. motion (se 3 ); By superimposing the various risk fields, a comprehensive risk field model R is constructed. total (se), calculate the security risk assessment value r(t) at the current time t based on the collected multi-dimensional situational data; Set risk threshold r max When the safety risk assessment value r(t) is greater than the risk threshold r max During this process, a driving decision-making process is conducted. Based on the current state, this process employs one or more decision-making combinations to reduce the safety risk assessment value r(t) to below the risk threshold r within a time interval Δt. max The level, wherein the decision includes: At least one first driving decision used to change the first situational element; At least one second driving decision used to alter the second situational element; And at least one third driving decision used to change the third situational element; The driving decision-making process includes the following steps: Determine the current driving status, driving status, and environmental status based on situational data. Based on the current state, select the decision types that can reduce the safety risk assessment value r(t) and construct the executable decision space A; Construct a decision model, wherein the decision model, within a time interval Δt, involves a series of decisions a = {a1, a2, ..., a...} N }, and a∈A, minimizing the decision value function is the optimization objective, where a1 is the initial decision; a N For the Nth decision; the expression for the optimization objective is: In the formula, r(t+1) is the safety risk assessment value at time t+1; the expression for the decision value function is: In the formula, V(t) is the decision value function at time t; V(t+1) is the decision value function at time t+1; and γ is the discount factor. Based on the optimization objective and the decision value function, determine one or more decision combinations within the time interval Δt.

2. The method for network security situation awareness and emergency response for autonomous driving according to claim 1, characterized in that, The first risk field model is a function of driving state, expressed as: R behavior (se 1 )=k1f(se 1 ); where k1 is a non-zero constant correction coefficient; f(se 1 ) represents the functional relationship between the first situational element and the driving state risk, reflecting the impact of driving state on the safety risk assessment value r(t). 1 This is the first situational data collected; The second risk field model is a function of the driving state, expressed as: R status (se 2 )=k2f(se 2 ); where k2 is a non-zero constant correction coefficient; f(se 2 The relationship between the second situational element and the driving state risk is represented by ), reflecting the impact of the driving state on the safety risk assessment value r(t). 2 This is the second situational awareness data collected; The third risk field model is a function of the environmental state, expressed as: R motion (se 3 )=k3f(se 3 ); where k3 is a non-zero constant correction coefficient; f(se 3 The relationship between the third situational element and environmental state risk is expressed as a function, reflecting the impact of environmental state on the safety risk assessment value r(t). 3 This is the second situational awareness data collected; The comprehensive risk field model is a superposition of various risk field models, where the expression for the superposition state is: R total (se)=R behavior (se 1 )+R status (se 2 )+R motion (se 3 ); The security risk assessment value is a functional relationship between the multi-factor situational data and the state risk at the current time t, and the calculation formula is: r(t) = k1f(se 1 )+k2f(se 2 )+k3f(se 3 ).

3. The method for network security situation awareness and emergency response for autonomous driving according to claim 2, characterized in that, The functional relationship between the r-th situational factors and the state risk f(se) r The process of obtaining ) includes the following steps: The state risk levels corresponding to the r-th situation element are classified into low risk, medium risk, and high risk according to the principle of low to high, and the risk values ​​of low risk, medium risk, and high risk are defined as α, respectively. r1 α r2 and α r3 And 0 < α r1 <α r2 <α r3 r < 1; r = 1, 2, 3; Several different levels of situational element data are labeled respectively; Construct a neural network model that takes the labeled r-th situational element data as input and the risk value corresponding to the data as output, and train the neural network model until its classification accuracy reaches the set expected value. The trained neural network model is used to obtain the risk value of the state risk corresponding to all r-th situation elements, and the functional relationship between the r-th situation element and the corresponding risk value is output as the functional relationship f(se) between the r-th situation element and the state risk. r ).

4. The method for network security situation awareness and emergency response for autonomous driving according to claim 3, characterized in that, The neural network model is a multilayer perceptron neural network model with one input layer, at least one hidden layer, and one output layer. The number of neurons in the input layer of the neural network model is equal to the number of input situational elements; the number of neurons in the output layer is 1; and the number of neurons in each hidden layer is not less than the average number of neurons in the input layer and the output layer.

5. The method for network security situation awareness and emergency response for autonomous driving according to claim 1, characterized in that, The process for determining the initial decision a1 includes the following steps: Acquire situational data reflecting the current state, and calculate the first risk value r1(t) reflecting the driving state risk, the second risk value r2(t) reflecting the driving state risk, and the third risk value r3(t) reflecting the environmental state risk based on the situational data, where r1(t) = k1f(se 1 ); r2(t)=k2f(se 2 ); r3(t)=k3f(se 3 ); Calculate the risk value r at the current time t. r (t) Contribution C to the safety risk assessment value r(t) r ( t );in, Take the maximum contribution C r(t) The corresponding r-th driving decision is taken as the initial decision space A1, where the initial decision space A1 contains several decision types that can reduce the safety risk assessment value r(t); In the initial decision space A1, such that [r(t)-r max ] 2 +[r(t+1)-r(t)] 2 The decision that takes the minimum value is taken as the initial decision a1.

6. A cybersecurity situational awareness and emergency response system for autonomous driving, characterized in that, The system is used to implement the steps of the autonomous driving network security situation awareness and emergency response method according to any one of claims 1-5, including: The data acquisition module is used to collect multi-dimensional situational awareness data in real time, including: The first data acquisition submodule is used to collect the first situational elements reflecting the driving status; The second data acquisition submodule is used to collect the second situational element reflecting the driving status; The third data acquisition submodule is used to collect the third situational element reflecting the environmental state; The risk model building module is used to construct risk field models based on collected situational data, including: Used to construct the first risk field model R based on the first situation element data. behavior (se 1 The first risk field model construction submodule; Used to construct the second risk field model R based on the second situation element data. status (se 2 The second risk field model construction submodule; Used to construct a third risk field model R based on third situation element data. motion (se 3 The third risk field model construction submodule; The integrated risk field model construction module is used to overlay various risk fields to construct an integrated risk field model R. total (se); The security risk assessment module is used to calculate the security risk assessment value r(t) at the current time t based on the collected multi-dimensional situational data. The driving decision-making module is used when the safety risk assessment value r(t) is greater than the set risk threshold r max During this process, a driving decision-making process is conducted, wherein, based on the current state, the driving decision-making process employs one or more decision-making combinations to reduce the safety risk assessment value r(t) to below the risk threshold r within a time interval Δt. max The levels include: The first driving decision generation submodule is used to formulate changes to the first situational element; The second driving decision generation submodule is used to formulate changes to the second situational factors; The third driving decision generation submodule is used to formulate changes to the third situational element; The driving decision execution module is used to execute the driving decision strategy generated by the driving decision formulation module.

7. The autonomous driving network security situation awareness and emergency response system according to claim 6, characterized in that, The system further includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, is capable of implementing the steps of the autonomous driving network security situation awareness and emergency response method according to any one of claims 1-5.

Images